Dynamic localization of SRSF1 in male mouse germ cells.

(A) Dynamic localization of SRSF1 during spermatogenesis. Co-immunostaining was performed using SRSF1 and γH2AX antibodies from adult mouse testes. DNA was stained with DAPI. Scale bar, 20 μm.

(B) Expression of Srsf1 in testes at different stages of development. The RT‒qPCR data were normalized to Gapdh. n=3.

(C) Western blotting of SRSF1 expression in testes at different stages of development. ACTB served as a loading control. The value in 16.5 dpc testes were set as 1.0, and the relative values of testes in other developmental periods are indicated. n=3.

(D) Localization and expression of SRSF1 in spermatogonia. Co-immunostaining was performed using PLZF and SRSF1 antibodies in 7 dpp, 14 dpp, and adult mouse testes. DNA was stained with DAPI. Arrowheads, spermatogonia. Scale bar, 50 μm

SRSF1-binding genes have an essential role in spermatogonia.

(A) Network showing GO enrichment analyses of SRSF1-binding genes.

(B) Representative genome browser views of spermatogonia-related gene transcripts bound by SRSF1. Higher peaks are marked by a lavender area.

(C) Localization and expression of the spermatogonia-related proteins in mouse testes. Scale bar, 5 μm.

SRSF1 plays critical roles in spermatogenesis and male fertility.

(A) Vasa-Cre mice were crossed with Srsf1Fl/Fl mice to generate Srsf1 cKO mice. Cre-mediated deletion removed exons 2, 3, and 4 of Srsf1 and generated a null protein allele.

(B) Genotyping PCR was performed using Vasa-Cre and Srsf1 primers.

(C) Co-immunostaining of SRSF1 and PLZF in 7 dpp Ctrl and cKO testes. DNA was stained with DAPI. Scale bar, 10 μm.

(D) Fertility test results showed a male infertility phenotype in the cKO mice (n= 5) compared to the Ctrl mice (n= 8). The number of pups per litter was determined in the cKO (n= 5) and Ctrl (n= 8) mice.

(E) Haematoxylin-eosin-stained epididymis sections from adult Ctrl and cKO mice were obtained. Scale bar, 100 μm.

(F) Cauda epididymal sperm counting was performed. n=3.

(G) Normal body weight in cKO mice. The body sizes and weights of adult Ctrl and cKO mice are shown as the mean ± SEM. n= 3.

(H) Testis atrophy in adult cKO mice. Testis sizes and weights of adult Ctrl and cKO mice are shown as the mean ± SEM. n= 5.

(I) Haematoxylin-eosin-stained testis sections from adult Ctrl and cKO mice were obtained. Scale bar, left panel: 200 μm, right panel: 100 μm. SC, Sertoli cell; SPG, Spermatogonium; SPC, spermatocyte; RS, round spermatid; ES, elongated spermatid.

Unpaired Student’s t-test determined significance; exact P value P ≥ 0.05, ****P < 0.0001. The points and error bars represent the mean ± SEM.

Loss of germ cells in adult cKO mouse testes.

(A) Co-immunostaining of PLZF and γH2AX in adult Ctrl and cKO testes. DNA was stained with DAPI. Scale bar, right panel: 25 μm, other panels: 100 μm.

(B) Co-immunostaining of VASA and TRA98 in adult Ctrl and cKO testes. DNA was stained with DAPI. Scale bar, right panel: 25 μm, other panels: 100 μm.

(C) Whole-mount co-immunostaining of TRA98 and SOX9 in adult Ctrl and cKO testes. DNA was stained with DAPI. White dashed lines, boundary of the tubule. Scale bar, right panel: 20 μm, other panels: 100 μm.

SRSF1 is required for spermatogonia survival.

(A) Testis sizes of 5 dpp, 7 dpp, and 14 dpp Ctrl and cKO mice are shown. The testis/body weight ratios (g/kg) of 5 dpp, 7 dpp, 14 dpp, and adult Ctrl and cKO mice are shown as the mean ± SEM. n= 4.

(B) Immunostaining of VASA in 5 dpp, 7 dpp, and 14 dpp Ctrl and cKO testes. DNA was stained with DAPI. Scale bar, 200 μm. Number of VASA-positive cells per tubule is the mean ± SEM. n= 3.

(C) TUNEL apoptosis assay was performed on sections from 7 dpp Ctrl and cKO testes. DNA was stained with DAPI. Scale bar, 20 μm. Rate of germ cell apoptosis is the mean ± SEM. n= 3.

Unpaired Student’s t-test determined significance; exact P value P ≥ 0.05, **P < 0.01, ****P < 0.0001. The points and error bars represent the mean ± SEM.

SRSF1 is required for homing of precursor SSCs.

(A) Co-immunostaining of VASA and SOX9 in 5 dpp Ctrl and cKO testes. DNA was stained with DAPI. Scale bar, 10 μm. Red dashed circles, tubule. White dashed circles, germ cell.

(B) The percentage of VASA positive basal cells is shown as the mean ± SEM. n= 4.

(C) Immunohistochemistry staining of FOXO1 in 5 dpp Ctrl and cKO testes. The nuclei were stained with haematoxylin. Scale bar, 10 μm. Arrowheads, FOXO1 in the nucleus. Arrows, FOXO1 in the cytoplasm. H, homing. AH, abnormal homing.

(D) The percentage of FOXO1 positive cells with nuclear FOXO1 is shown as the mean ± SEM. 913 Ctrl germ cells and 592 cKO germ cells were obtained from 4 animals.

(E) The percentage of FOXO1 positive basal cells with nuclear FOXO1 is shown as the mean ± SEM. 913 Ctrl germ cells and 592 cKO germ cells were obtained from 4 animals.

Unpaired Student’s t-test determined significance; ***P < 0.001, ****P < 0.0001. The points and error bars represent the mean ± SEM.

SRSF1 regulates the expression of spermatogonia-related genes.

(A) Expression of Srsf1 in 5 dpp Ctrl and cKO mouse testes. The RT‒qPCR data were normalized to Gapdh. n=5. The expression of Srsf1 is shown as reading coverage in 5 dpp mouse testes.

(B) Western blotting of SRSF1 expression in 5 dpp mouse testes. ACTB served as a loading control. The value in Ctrl testes was set as 1.0, and the relative values in cKO testes are indicated. n=5.

(C) Volcano map displaying the distribution of differentially expressed genes from RNA-seq data. The abscissa in the figure represents the gene fold change in 5 dpp cKO and Ctrl mouse testes. |log2FoldChange| ≥ 0. The ordinate indicates the significance of gene expression differences between 5 dpp cKO and Ctrl mouse testes. padj ≤ 0.05. Upregulated genes are shown as red dots, and downregulated genes are shown as green dots.

(D) Cluster heatmap of differentially expressed genes. The ordinate is the genotype, and the abscissa is the normalized FPKM (fragments per kilobase million) value of the differentially expressed gene. Red indicates a higher expression level, while blue indicates a lower expression level.

(E) Network showing GO enrichment analyses of differentially expressed genes.

(F) Heatmap of spermatogonia-related gene expression.

(G) The expression of spermatogonia-related genes is shown as read coverage.

(H) The expression of spermatogonia-related genes in 5 dpp cKO and Ctrl mouse testes. The RT‒qPCR data were normalized to Gapdh. The value in the Ctrl group was set as 1.0, and the relative value in the cKO group is indicated. n=3. Unpaired Student’s t-test determined significance; exact P value P ≥ 0.05, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. The points and error bars represent the mean ± SEM.

SRSF1 directly binds and regulates the expression and AS of Tial1/Tiar.

(A) Venn diagram showing the correlation among down-regulated, upregulated, alternatively spliced, and SRSF1-binding genes.

(B) Schematic diagram showing the classes of splicing events.

(C) Splicing events were analysed by number, exclusion, and inclusion.

(D) Network showing GO enrichment analyses of AS genes.

(E) The AS of Tial1/Tiar in 5 dpp Ctrl and cKO mouse testes was analysed by RT–PCR. n=3. The ratio of inclusion (Incl) to exclusion (Excl) is shown accordingly.

(F) Analyses of Tial1/Tiar expression and exon‒exon junctions were performed.

(G) SRSF1 directly binds the pre-mRNA of Tial1/Tiar by RIP–qPCR in 5 dpp mouse testes. n=3.

(H) The FPKM of Tial1/Tiar in 5 dpp Ctrl and cKO mouse testes.

(I) The expression of Tial1/Tiar in 5 dpp Ctrl and cKO mouse testes. The RT‒qPCR data were normalized to Gapdh. The value in the Ctrl group was set as 1.0, and the relative value in the cKO group is indicated. n=3.

(J) Western blotting of TIAL1/TIAR expression in 5 dpp mouse testes. ACTB served as a loading control. The FL/isoform X2 value of TIAL1/TIAR in the Ctrl group was set as 1.0, and the relative value in the cKO group is indicated. n=3. FL, full length.

Unpaired Student’s t-test determined significance; exact P value P ≥ 0.05, ***P < 0.001, ****P < 0.0001. The points and error bars represent the mean ± SEM.

SRSF1 recruits AS-related proteins to modulate AS in testes.

(A) Silver-stained gel of SRSF1 and control immunoprecipitates from 5 dpp mouse testis extracts.

(B) IP experiment was performed in 5 dpp mouse testis extracts.

(C) IP of SRSF1 from IP-MS data.

(D) Pearson’s correlation analysis showed the coefficient between the two samples for IP-MS data.

(E) Network showing GO enrichment analyses of SRSF1-binding proteins.

(F) Circular heatmap of AS-related proteins.

(G) Co-IP experiment for detecting the SRSF1 association between SART1, RBM15, and SRSF10 in 293T cells.

(H) SRSF1-mCherry cotransfected with SART1-eGFP, RBM15-eGFP, and SRSF10-eGFP 293T cells is shown. DNA was stained with DAPI. Scale bar, 10 μm.

(I) Co-IP experiment for detecting the SRSF1 truncated protein association between SART1, RBM15, and SRSF10 in 293T cells.

(J) A schematic diagram of protein interactions is shown.

(K) Schematic illustration of the molecular mechanisms by which SRSF1 regulates homing of mouse precursor SSCs.

Dynamic localization of SRSF1 during spermatogenesis.

Co-immunostaining was performed using SRSF1 and γH2AX antibodies in adult mouse testes. DNA was stained with DAPI. Scale bar, 40 μm.

The libraries of RNA-seq are of good quality.

(A) Filtering of sequencing data per sample.

(B) Distribution of sequenced reads across genomic regions for all samples.

(C) Gene expression values (FPKM) of all samples were subjected to principal component analysis (PCA).

(D) The distribution of gene expression levels of different samples is demonstrated by violin plots as shown below.

The horizontal coordinate of the graph is the name of the sample, and the vertical coordinate is log2 (FPKM+1).

Transcription and translation information for Tial1.

(A) Transcription and translation information for Tial1 was obtained from ensembl genome browser.

(B) Transcription and translation information for Tial1 was obtained from genome data viewer.